Use dbt_artifacts

[ataft]

This project should import the https://github.com/brooklyn-data/dbt_artifacts package and then make necessary improvements like model rowcount, adding the columns model, column stats, and other small things like model description.

The dbt_artifacts project now supports postgres/redshift, and has a lot of improvements like chunking inserts when there are a lot of models.